Abstract This Phase I project will focus on developing key elements needed to achieve a wearable, high-density, magnetoencephalography (MEG) device based on optically-pumped atomic magnetometers (OPMs). OPM sensors have progressed to be comparable in sensitivity to liquid-helium-cooled superconducting sensors (SQUIDs) but without the complexity and bulk required by cryogenic cooling. An OPM-based MEG provides further advantages such as lower cost and the ability to place sensors directly on the subject?s head. It has recently been shown that large, non-invasive ?on-scalp? arrays of OPMs (1) will outperform traditional fixed helmet MEG devices by a factor of 7.5 higher signal, and (2) may reach the same resolution as invasive Electrocorticography (ECoG). However, there are a few advancements that must be made to evolve from the proof-of-concept systems of a few tens of sensors to a practical full head MEG system capable of producing high resolution images. In this project will address challenges like cross-talk between sensors, background field compensation, and sensor calibration and localization. We believe these innovations are necessary for a fully-integrated, wearable, high-density, and on-scalp MEG device using ultra-sensitive OPMs.